Skin?s function as a barrier to infection, dehydration and chemical assault is critical to health and survival. Its barrier effectiveness rests almost entirely in the thin, outer membrane, called the stratum corneum (SC), which consists of dead skin cells embedded in a highly organized dense lipid-rich environment. This organization of the SC lipids into ordered gel or crystalline phases can be ascribed to its unique composition: mostly ceramides, free fatty acids and cholesterol, with no phospholipids in contrast to most biological membranes. There is compelling evidence that an impaired skin barrier, which is coincidental with abnormalities in composition, organization and structure of the SC lipids, is the primary event in the pathogenesis of skin disease and even some systemic diseases (e.g., the occurrence of asthma and allergic rhinitis in patients with atopic dermatitis). In contrast, the effectiveness of the SC barrier is a central problem limiting topical and transdermal drug delivery. An improved understanding of the relationship between SC lipid composition, structure, and organization and barrier function is needed: (1) to understand the relationship between skin disease, reduced barrier function, and effective treatment, and (2) to develop new techniques for either reducing barrier function (to deliver drugs more effectively) or improving it (to protect against disease, toxic exposure and water loss). While experimental lipid systems that mimic the SC can be designed and studied, any understanding of the relationship between barrier function and lipid composition and organization can only be inferred; accurate computational studies on well-characterized systems would allow the mechanistic basis of these relationships to be clearly probed. To complement existing model lipid systems that mimic healthy SC, we will experimentally design and characterize new model lipid systems that mimics diseased skin, specifically atopic dermatitis (AD). These well-characterized synthetic lipid models will form the basis of the systems to be explored computationally. Using a multi-scale modeling approach that seamlessly combines simulations at the atomistic and coarse-grained (CG) levels, unprecedented insight into the molecular level organization of, and interactions between, SC lipid molecules in equilibrated assemblies of SC lipids modeling normal and AD-SC will be obtained. Targeted experiments will also be performed to provide the data needed to validate the model predictions (structures of the simulated lipid assemblies will be compared with spectroscopic, scattering and permeability studies on the corresponding experimental systems) before the computational framework is used for predictive in silico screening to probe various hypothesis related to barrier function and determine the sensitivity of the SC structure and thus barrier function to changes in lipid composition. In contrast to experimental studies, the computational models allow for variations in the composition of the SC models to be easily adjusted. The work is unique since self-assembled structures will form the basis of the modeling studies and a synergistic experimental and computational multiscale approach will be employed.